8o SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 8o 



reason for believing that in the dragonflies the two parts of the tro- 

 chanter (iTr, 2Tr) represent two primitive segments of the leg, as 

 claimed by Verhoeff (1903 a, 1903 b). Griindberg (1903), on the 

 other hand, sees here only a secondary division of the trochanter, 

 and he would homologize the dividing ridge {h) with the more basal 

 ridge of the trochanter in other insects. He fails to note, however, 

 that the latter ridge {g) is also present and well developed in the 

 Odonata, and bears the coxal articulations (/). Verhoeff calls the 

 first trochanter the " true trochanter " and the second the " prae- 

 femur." He believes that the first disappears in most insects other 

 than the Odonata, and that the so-called trochanter of insects is the 

 homologue of the praefemur of the Chilopoda. The disappearance 

 of the first trochanter segment, however, would involve a replace- 

 ment of the articular elements in the coxo-trochanteral joint, an un- 

 likely transformation. It seems more reasonable, therefore, that the 

 two trochanteral segments have been united into one in most insects 

 (fig. 42). The basal lip of the trochanter (fig. 35 A), when un- 

 usually wide, often gives the trochanter a false appearance of being 

 a double segment. 



The femur. — This, the third segment of the insect leg (fig. 31 A, 

 F), is usually the longest and strongest part of the limb; but it 

 varies in size from that of the huge hind femur of leaping Orthoptera 

 to that of the small femoral segment in the leg of a sawfly larva 

 (fig. 16 B), which is much inferior to the coxa. The femoro-tibial 

 joint, or " knee " of the insect leg, is typically dicondylic in adult 

 insects and in the nymphs of Hemimetabola ; but in holometabolous 

 larvae it is usually monocondylic (figs. 16 A, 32 A), as it is in the 

 Chilopoda (fig. 32 B). In the larvae of Neuroptera and Trichoptera 

 it is dicondylic. 



The tibia. — The tibia (fig. 31 A, Th) is characteristically a slender 

 segment in adult insects, only a little shorter than the femur, or the 

 combined femur and trochanter. Its proximal end forms a more or 

 less distinct " head " bent toward the femur, a device which allows 

 the tibia to be flexed close against the ventral surface of the femur, 

 and one often not expressed sufficiently in insect drawings to suggest 

 the essential mechanism of the femoro-tibial articulation. The tibio- 

 tarsal joint is dicondylic in adult insects, unless articular points are 

 lacking, but it is always monocondylic in holometabolous larvae. 

 Sometimes the tibia and tarsus are united, forming a tibio-tarsal 

 segment (figs. 16 B, 43 B). 



The tarsus. — In adult insects, the tarsus comprises from one to 

 five small pieces (fig. 31 A, Tar). In holometabolous larvae, however, 



